Ebook: Ontologies and Semantic Technologies for Intelligence

This book had its origin at the Second International Ontology for the Intelligence Community (OIC) Conference, which was held on November 28–29, 2007, in Columbia, MD, USA. At that time, a volume was proposed by the editors that would feature chapters by selected authors from the conference, who could extend their OIC papers or write on related topics that fit the guidelines the editors established for this book. In addition, other authors were invited to submit chapters.

This book represents a partial technology roadmap for government information technology decision makers for information integration and sharing, and situational awareness (improved analysis support) in the use of ontologies, and semantic technologies for intelligence.

The general themes of both the OIC conferences and this book focus on intelligence community needs and the applications of ontologies and semantic technologies to assist those needs. Among the very many IC needs are the following:

•To increase the ability to meaningfully share information, within and among communities, across humans and machines

•To off-load some human cognitive functions and enable machines to assume these. By using ontologies and semantic technologies, machines come up to the human conceptual level, rather than humans having to go down to the machine level, which latter tack has largely defined information technology since its orgins up to this point.

•To increase the ability to automate some aspects of intelligence analysis, as for example, by supporting evidence-based reasoning, deductive (what logically follows, given the knowledge) and abductive (what is the best explanation, given the evidence) queries

•To provide assistance on probability of Hypothesis given the Evidence P(H|E), hypothesis generation, and analysis of competing hypotheses by using complex knowledge and logical mechanisms, and evaluating the consequences or ramifications of hypotheses

•To increase the capability to semantically integrate data from all intelligence disciplines

•To provide analytical tools that exploit the availability of semantically integrated information and knowledge

•To assist in semantic disambiguation, reference, co-reference/correlation of entities, relations, and events

○Disambiguation: To determine the appropriate meaning for the given context

○Reference: To determine the actual entity in the world that the data refers to

○Co-reference/correlation: To determine whether two entities are actually the same entity, and the properties and events those entities respectively possess and participate in

•To support reasoning over geospatial, temporal, and other data to infer additional information about the real world.

The target audience of this book is the US and other intelligence communities (IC), including law enforcement and homeland security communities, along with other technical and budgetary decision makers and technologists working in intelligence. These technologists include ontologists and ontology developers, computer scientists, software engineers, and intelligence analysts who have a strong interest in semantic technologies and their applications.

This book would not have been possible without the assistance, dedication, and patience of many generous individuals. We thank the IOS Press publisher and its dedicated representive Maarten Fröhlich for tolerance of delays in the editing of this book, while also providing constant and continuing encouragement. We thank the very many anonymous reviewers who helped improve the contributions of the authors by offering sound feedback and critical comments on multiple iterations of chapters. We thank the past organizers of the OIC conferences, for valuable suggestions and help on many issues, including in particular Barry Smith, Kathryn Blackmond Laskey, Duminda Wijesekera, and Paulo Cesar G. da Costa. We also thank Kevin Lynch and David Roberts, who provided governmental support for the OIC conferences and also feedback to the authors and editors on the impact of these technologies on the intelligence community, thereby serving to provide a pragmatic perspective to constrain the potential technological exuberance. Of course the editors also thank their friends and families, who have countenanced aggravation, missed social opportunities, and personal inattention, to enable the writing and editing of this volume.

Finally, we underscore that the views expressed in this book are those of the authors alone and do not reflect the official policy or position of The MITRE Corporation, the Lockheed-Martin Corporation, or any other company or individual, nor that of any particular intelligence community, agency, organization, or government.

Leo Obrst,

Terry Janssen,

Werner Ceusters

In recent years ontologies and semantic technologies more generally have begun to be applied to assist the intelligence community, for information integration, information-sharing, decision-support, and in many other applications. This chapter introduces the topic of the book and provides background information concerning its rationale, historical perspective, a vision for the future, and briefly describes the chapters of the present volume.

The analysis of events prior to and during September 11 revealed that a smooth execution of the intelligence process is hampered by inadequate information sharing. This caused a rethinking of the intelligence process and a transition towards a ‘Globally Networked and Integrated Intelligence Enterprise’ with the goal that more detailed, tagged, and, therefore, traceable, information will reach those who need it, when they need it, and in a form that they can easily absorb. We present the referent tracking paradigm and its implementation in networks of referent tracking systems as an enabling technology to make this vision come true. Referent tracking uses a system of singular and globally unique identifiers to track not only entities and events in first-order reality, but also the data and information elements that are created to describe such entities and events in information systems. By doing so, it meets the requirements of the Nation's Information Sharing Strategy.

The blogosphere provides a novel window into an important segment of public opinion, but its dynamic nature makes it an elusive medium to analyze and interpret in the aggregate, where it is most informative. We are developing a new open-source blog mining technology that employs ontologies to solve this problem by fusing the signals of the blogosphere and zeroing in on issues that are most likely to migrate offline. This technology is designed to enable analysts to anticipate the threats or opportunities these issues represent in a timely and efficient fashion.

Lockheed Martin Corp. has funded research to generate a framework and methodology for developing semantic reasoning applications to support the discipline of Intelligence Analysis. This chapter outlines that framework, discusses how it may be used to advance the information sharing and integrated analytic needs of the Intelligence Community, and suggests a system / software architecture for such applications.

Ontologies enable explicit expression of collective concepts and support Machine-to-Machine (M2M) interactions at the semantic level. Ontologies expressed in a standard language, such as the Web Ontology Language (OWL) and exposed on a network offer the potential for unprecedented interoperability solutions since they are semantically rich, computer interpretable and inherently extensible. In this chapter, we describe how we applied ontologies in OWL for rapid enterprise integration of heterogeneous data sources to track objects in a battlespace. We found that once a robust foundational domain ontology is established, it is easy and quick to integrate new data sources and therefore rapidly provide new system capabilities. In particular, we demonstrate how moving tracks can be quickly integrated with intelligence and space events to provide enhanced situational awareness using ontologies. This chapter also describes the overall SEER and SWORIER systems we developed, the latter of which translated OWL ontologies (and RDF instances) and Semantic Web Rule Language (SWRL) rules into Prolog, applied knowledge compilation techniques, and then at runtime, utilized a combined OWL/logic programming reasoned for efficient automated reasoning. We also briefly describe a more recent extension to the prototype and to the ontologies that we made to address more rigorous geospatial rules for unmanned autonomous vehicle (UAV) avoidance. Finally, we consider some issues raised by our work and future lines of research to address these.

This paper presents a new paradigm for imagery analysis where imagery is annotated using terms defined in ontologies, enabling more powerful querying and exploitation of the analysis results. The ontology terms represent the concepts and relationships necessary to effectively describe the objects and activities within a domain of interest. A platform for viewing and editing imagery annotations is described along with a specialized semantic knowledge base capable of efficiently querying the information using semantic, spatial, and temporal qualifiers. The ontologies used for representing the annotations and domain of interest are also described.

We describe provability-based semantic interoperability (PBSI), a framework transcending syntactic translation that enables robust, meaningful, knowledge exchange across diverse information systems. PBSI is achieved through translation graphs that capture complex ontological relationships, and through provability-based queries. We work through an example of automating an unmanned aerial vehicle by reasoning over information from a number of sources.

For some years now, the Intelligence Community has been using XML “tagging” of documents in an effort to make the documents more usable for data discovery, sharing, and other processing. In implementing a system (METS) which automates the identification of relevant data in documents, we noted several limitations of that XML tagging approach, and therefore chose to also provide an OWL ontology-based representation of that data. Here, we discuss the goals of METS, those XML limitations, and the OWL approach, showing how the latter should support better analysis. (As we discuss, clients have thus far not made use of the OWL results, so the benefits are still hypothetical.) We also discuss issues we encountered in developing the ontologies, outline the design and use of the operational METS for processing message feeds and other documents, and conclude with future plans which include greater ontology coordination and sharing, and assisting with the incorporation of tools for benefiting from the semantic information provided by the OWL.

Systems are increasingly required to fuse data from geographically dispersed, heterogeneous information sources to produce up-to-date, mission-relevant results. These products focus not only on traditional military forces and systems, but to an increasing degree also on non-traditional combatants and their social networks. Successful multi-INT fusion requires that the constituent systems interoperate not just at the level of syntax and formats, but also at the level of semantics. Ontologies are vital enablers for semantic interoperability. Because uncertainty is a fundamental aspect of multi-INT fusion, lack of support for uncertainty is a major limitation of current-generation ontology formalisms. Probabilistic OWL (PR-OWL) extends the OWL Web Ontology Language to enable the construction of probabilistic ontologies. Ontologies constructed in PR-OWL can represent complex patterns of evidential relationships among uncertain hypotheses. Recently, a system for specifying and reasoning with PR-OWL ontologies has been released in beta version. This paper describes the PR-OWL ontology language, the probabilistic logic on which it is based, and the reasoning system implementation. A hypothetical case study in the counterterrorism domain illustrates the capabilities of PR-OWL.

This chapter describes work on an integrated system that can assist analysts in exploring hypotheses using Bayesian analysis of evidence from a variety of sources. The hypothesis exploration is aided by an ontology that represents domain knowledge, events, and causality for Bayesian reasoning, as well as models of information sources for evidential reasoning. We are validating the approach via a tool, Magellan, that uses both Bayesian models and logical models for an analyst's prior knowledge about how evidence can be used to evaluate hypotheses. The ontology makes it possible and practical for complex situations of interest to be modeled and then analyzed formally.

It has been estimated that up to 80% of all information contains some notion of location. This is helping create a greater understanding of the utility of geospatial information as a framework for organizing, portraying and better understanding other information and the relationships of people, places, things and events. Geospatial capabilities are entering the mainstream of information technology and spatial data infrastructures (SDI's) are being implemented to bring together the technologies, policies, standards, and human resources to better utilize geospatial data. SDI's such as the National System for Geospatial-Intelligence (NSG) are using a standards baseline of ISO, Open Geospatial Consortium and other relevant consensus standards and putting service oriented architectures in place to achieve distributed, data-centric, net-centric operations. This stage of development of SDI's is bringing an unprecedented level of interoperability to geospatial data and technology and is setting the stage for an even greater level of future interoperability and data integration. The development of geospatial ontologies and semantic capabilities for integrating well structured geospatial data with unstructured geospatial information existing in other data sets will be the catalyst for this next major step forward.

The National Geospatial-Intelligence Agency is in the forefront of examining the current state, future potential and implementation requirements for a semantically enabled geospatial web. This Geospatial Ontology Trade Study is a broad survey of ontologies. An ontology is a formal, explicit, shared conceptualization of a domain and defines the concepts and vocabulary used within a community of interest. The study report outlines the characteristics of the ontologies surveyed and makes recommendations about which are best suited for certain types of uses and identifies further research and work to formalize geospatial ontologies.

The report concludes that there are a number of existing standards-based ontologies which provide building blocks for geospatial representations and makes recommendations for strategic actions to incorporate ontologies and semantic knowledge into the growing base of Geospatial Intelligence capabilities.

This chapter looks at the intersection of intelligence and ontologies and semantic technologies, and tries to characterize the impact of these in the future. It provides a view into some emerging technologies such as query languages and rule standards for the Semantic Web. It also provides some guidance from a different domain, the biomedical domain, and tries to show that realist ontologies, ontologies based on common real world characterizations, have an effective impact on applications in those domains. Finally, it looks at the potential impact of these technologies on intelligence collection and analysis in the future, and makes some predictions.